About
Hi! I am an incoming PhD student at The Chinese University of Hong Kong supervised by Prof. Wai Lam. I am currently a senior at Harbin Institute of Technology, Weihai, advised by Prof. Dianbo Sui, and interning at PCG, Tencent, advised by Dr. Qingbin Liu. I have also had the privilege of working as a research assistant and visiting student at HKUST(GZ) and HKU CDS.
My research interests span Large Language Models (LLMs) — particularly Model / Knowledge Editing, Agent, and Efficient ML — and more recently, Diffusion Models,Video Generation and World Models.
Before pivoting to research, I was an active competitive programmer and served as leader of the ACM club, mentored by Prof. Kaikun Dong.
I am always happy to discuss research ideas and explore potential collaborations!
Open-Source Projects
ForgeWM
First open-source pipeline to train real-time, keyboard & mouse controllable world models.
A reproducible training recipe for action-controllable world models. An integration project connecting Matrix-Game 2's game-native I2V backbone, GameFactory's open Minecraft data, and the Causal Forcing distillation paradigm into the first fully open, end-to-end pipeline for the MG2 lineage — trainable on 8 GPUs.
Selected Publications
View All →
ScEdit: Script-based Assessment of Knowledge Editing
Xinye Li, Zunwen Zheng, Qian Zhang, Dekai Zhuang, Jiabao Kang, Liyan Xu, Qingbin Liu, Xi Chen, Zhiying Tu, Dianhui Chu, Dianbo Sui
Findings of the Association for Computational Linguistics: ACL 2025, 2025
Script-based assessment for knowledge editing in LLMs.
Cognitive Alpha Mining via LLM-Driven Code-Based Evolution
Fengyuan Liu, Huang Yi, Sichun Luo, Yuqi Wang, Yazheng Yang, Xinye Li, Zefa Hu, Junlan Feng, Qi Liu
The 64th Annual Meeting of the Association for Computational Linguistics, 2026
Cognitive Alpha Mining via LLM-Driven Code-Based Evolution. Work done during the internship at GIM AI.
Exploring Deductive and Inductive Reasoning Capabilities of Large Language Models in Procedural Planning
Jiabao Kang, Xinye Li, Liyan Xu, Qingbin Liu, Xi Chen, Zhiying Tu, Dianhui Chu, Dianbo Sui
Findings of the Association for Computational Linguistics: EMNLP 2025, 2025
Exploring deductive and inductive reasoning capabilities of LLMs in procedural planning tasks.
Can We Debias Multimodal Large Language Models via Model Editing?
Zecheng Wang, Xinye Li, Zhanyue Qin, Chunshan Li, Zhiying Tu, Dianhui Chu, Dianbo Sui
Proceedings of the 32nd ACM International Conference on Multimedia (ACM MM), 2024
Investigating debiasing multimodal LLMs through model editing techniques.
News
🌍 ForgeWM has been released! The First open-source pipeline to train real-time, keyboard & mouse controllable world models.
🏆 Awarded the 🎓 CUHK Vice-Chancellor's Ph.D. Scholarship!
🎉 One Paper has been accepted by 🇺🇸 ACL 2026 KnowFM Workshop!
🎉 Two papers (Vid-LLMs Editing & CogAlpha) have been accepted by Main Conference of 🇺🇸 ACL 2026!
🎉 I got the Phd offer from 🇭🇰 The Chinese University of Hong Kong (CUHK)!
🏆 Awarded the Zeshi Virtue and Talent Scholarship, one of the prestigious scholarships at Harbin Institute of Technology, Weihai!
🏆 Honored to be awarded the National Scholarship 2024-2025 (Top 0.4% Nationwide)!
🎉 Our paper on deductive and inductive reasoning has been accepted by Findings of 🇨🇳 EMNLP 2025!
🎉 Our paper ScEdit has been accepted by Findings of 🇦🇹 ACL 2025! My first first-author paper!
🎉 Shared task paper LLMSR@XLLM25 accepted by ACL 2025 XLLM Workshop.
🥉 Our team won a Bronze Medal in ICPC EC-Final!
🏅 Our project was awarded Second Prize in the "Challenge Cup".
🎉 Our paper on debiasing MLLMs via model editing has been accepted by 🇦🇺 ACM MM 2024!
Honors and Awards
- 2026.05, CUHK Vice-Chancellor's Ph.D. Scholarship The Chinese University of Hong Kong.
- 2025.12, Zeshi Virtue and Talent Scholarship (10 in HITWH), Harbin Institute of Technology, Weihai.
- 2025.12, National Scholarship (Top 0.4% Nationwide), Ministry of Education of China.
- 2024.12, "Excellent Student Model" Honorary Title (Top 0.5%), Harbin Institute of Technology.
- 2023.12, National Scholarship (Top 0.2% Nationwide), Ministry of Education of China.
- 2023.11, Taihu Future Science and Technology Scholarship (Top 1.6%), Harbin Institute of Technology, Weihai.
- 2023.04, First Grade Scholarship Harbin Institute of Technology, Weihai.
Competitions
- 2024.12: 49th ACM-ICPC Asia East Continent Final (EC-Final) (Bronze Medal, Team Member)
- 2024.11: 19th "Challenge Cup": 2024 'Open Call for Solutions' Special Program (Second Prize, Team Leader)
- 2024.09: 2024 Weihai Collegiate Programming Contest (Gold Medal, Team Member)
- 2024.04: Huawei CodeCraft Contest 2024 (Second Prize, 9th in Preliminary Round)
- 2023.12: 48th ACM-ICPC Asia Regional Contest, Hangzhou (Silver Medal, Team Member)
- 2023.11: 48th ACM-ICPC Asia Regional Contest, Nanjing (Bronze Medal, Team Member)
- 2023.10: 2023 CCF Collegiate Computer Systems & Programming Contest (Bronze Medal, Team Member)
Education
- 2026.08 - 2030.07 (Expected), The Chinese University of Hong Kong. Incoming PhD Student.
- 2022.08 - 2026.06, Harbin Institute of Technology, Weihai, China. B.Eng. in Software Engineering.
- 2025.04 - 2025.10, Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China. Visiting Research Student.
- 2019.09 - 2022.06, Dalian No.24 High School, Dalian, China. High School Diploma.
Internships
Services
- ACM (Competitive Programming) Club Leader @ HITWH.
- Reviewer: Open to reviewing opportunities in AI-related workshops and venues.
Misc
- Several Useful Information about Summer Intern & PhD Application (Mainly focused on HK).
- If you'd like to collaborate or want advice on applying to CS PhD/MPhil programs, feel free to email me directly.
- I run a RedNote account, XIEEE, where I share graduate school application tips and research experiences.
- I've been learning Cantonese on Duolingo for 200+ days, but I'm still "識聽唔識講" (may understand but struggle to speak).